Mon. Not. R. Astron. Soc. 000, 1{27 (2002) Printed 3 April 2017 (MN LATEX style file v2.2) Predictions for the secondary CO, C and O gas content of debris discs from the destruction of volatile-rich planetesimals Quentin Kral?, Luca Matr`a,Mark C. Wyatt, Grant M. Kennedy Institute of Astronomy, University of Cambridge, Madingley Road, Cambridge CB3 0HA, UK Accepted 1928 December 15. Received 1928 December 14; in original form 1928 October 11 ABSTRACT This paper uses observations of dusty debris discs, including a growing number of gas detections in these systems, to test our understanding of the origin and evolution of this gaseous component. It is assumed that all debris discs with icy planetesimals create second generation CO, C and O gas at some level, and the aim of this paper is to predict that level and assess its observability. We present a new semi-analytical equivalent of the numerical model of Kral et al. (2016) allowing application to large numbers of systems. That model assumes CO is produced from volatile-rich solid bodies at a rate that can be predicted from the debris discs fractional luminosity. CO photodissociates rapidly into C and O that then evolve by viscous spreading. This model provides a good qualitative explanation of all current observations, with a few exceptional systems that likely have primordial gas. The radial location of the debris and stellar luminosity explain some non-detections, e.g. close-in debris (like HD 172555) is too warm to retain CO, while high stellar luminosities (like η Tel) result in short CO lifetimes. We list the most promising targets for gas detections, predicting > 15 CO detections and > 30 CI detections with ALMA, and tens of CII and OI detections with future far-IR missions. We find that CO, CI, CII and OI gas should be modelled in non-LTE for most stars, and that CO, CI and OI lines will be optically thick for the most gas-rich systems. Finally, we find that radiation pressure, which can blow out CI around early-type stars, can be suppressed by self-shielding. Key words: accretion, accretion discs hydrodynamics interplanetary medium plan- etdisc interactions circumstellar matter Planetary Systems. 1 INTRODUCTION Molecular CO gas is observed in the sub-mm with both arXiv:1703.10693v1 [astro-ph.EP] 30 Mar 2017 single-dish telescopes (JCMT, APEX) and interferometers Gas is observed around a growing number of planetary sys- such as ALMA, the SMA or NOEMA. For the brightest tar- tems where planets are likely to be formed and the proto- gets, ALMA's high-resolution and unprecedented sensitivity planetary discs in which they formed are already gone. All allow us to obtain CO maps for different lines and isotopes these gas detections are in systems where secondary dust showing the location of CO belts and giving an estimate of is created from collisions by bigger bodies orbiting in a de- their mass (see the CO gas disc around β Pic, Dent et al. bris belt similar to the Kuiper or asteroid belt in our so- 2014; Matr`aet al. 2017). Atomic species are also detected lar system. Similarly, the observed gas around these mature around a few debris disc stars. In particular, Herschel was systems may also be of secondary origin and being released able to detect the OI and CII fine structure lines in two from debris belt planetesimals/dust owing to grain-grain col- and four systems, respectively (e.g. Riviere-Marichalar et lisions (Czechowski & Mann 2007), planetesimal breakup al. 2012; Roberge et al. 2013; Riviere-Marichalar et al. 2014; (Zuckerman & Song 2012), sublimation (e.g. Beust et al. Cataldi et al. 2014; Brandeker et al. 2016). Also, metals have 1990), photodesorption (Grigorieva et al. 2007) or giant im- been detected, using UV/optical absorption lines, around β pacts (Lisse et al. 2009; Jackson et al. 2014). For some sys- Pictoris (Na, Mg, Al, Si, and others, Roberge et al. 2006), tems such as HD 21997, the observed gas may be of primor- 49 Ceti (CaII, Montgomery & Welsh 2012), and HD 32297 dial origin (K´osp´alet al. 2013). (NaI, Redfield 2007). Some of these metals are on Keplerian orbits but should be blown out by the ambient radiation pressure (Olofsson et al. 2001). It is proposed that the over- ? E-mail: [email protected] c 2002 RAS 2 Q. Kral abundant ionised carbon observed around β Pic, which is take the local interstellar radiation field derived by Draine not pushed by radiation pressure could brake other ionised 2011), in which case the detectability of gas in the model species due to Coulomb collisions with them (Fern´andez et depends only on R0, M_ CO,L? and d. We will explain which al. 2006). A stable disc of hydrogen has not yet been ob- parts of this parameter space should be preferentially ob- served in these systems (Freudling et al. 1995; Lecavelier served when trying to detect CO, CI, CII or OI with dif- des Etangs et al. 2001) but some high velocity HI compo- ferent instruments. This will give a general understanding nent (presumably falling onto the star) was detected recently of gas observations in debris discs and is particularly well with the HST/COS around β Pic (Wilson et al. 2017). All suited for planning mm-wave APEX/ALMA line observa- these observations need to be understood within the frame- tions and considering the science that could be done with work of a self-consistent model. Models of the emission of future missions such as SPICA (Swinyard et al. 2009) or the gas around main sequence stars have been developed, NASA's far-IR surveyor concept (FIRS, now called the Ori- but gas radial profiles were not derived self-consistently and gins Survey Telescope) that may be built within the next 15 often assumed to be gaussian (e.g., Zagorovsky et al. 2010) years. or not to be depleted in hydrogen compared to solar (as In section 2, a summary of gas detections around nearby expected in debris discs, e.g., Gorti & Hollenbach 2004) or main sequence stars is presented. In section 3, we present CO both (e.g., Kamp & Bertoldi 2000). abundance predictions for a large sample of debris discs and One self-consistent model has been proposed in Kral explain which systems are more likely to have CO detected. et al. (2016) (KWC16) that can explain gas observations In section 4, we present a similar analysis for CI and CII around β Pictoris. It proposes that CO gas is released from and go on with OI predictions in section 5. We discuss our solid volatile-rich bodies orbiting in a debris belt as first findings in section 6 before concluding in section 7. proposed by Mo´oret al. (2011); Zuckerman & Song (2012), and verified by Dent et al. (2014); Matr`aet al. (2017). CO is then photodissociated quickly and produces atomic carbon 2 GAS OBSERVATIONS IN DEBRIS DISCS and oxygen gas that evolves by viscous spreading, param- eterised with an α viscosity, resulting in an accretion disc The number of debris disc systems with gas detected is grow- inside the parent belt and a decretion disc outside. A steady ing and we now have 12 systems that can help us to under- state is rapidly reached (on a viscous timescale), meaning stand the dynamics of this gas and its origin. These systems that it is unlikely that we observe a system in a transient are presented in Table 1. Ten of them have CO detections, phase. The α viscosity could be provided by the magneto- whilst 2 have CI detected, 4 have CII detected and 3 systems rotational instability (MRI) as presented in Kral & Lat- have OI detected. All of these systems are shown in Fig. 1 ter (2016). Gas temperature, ionization state and popula- in a L? versus R0 diagram (see Table 1 to find the values tion levels are computed using the photodissociation region used and their references). (PDR) model Cloudy at each time step (Ferland et al. 2013). In this plot, we show the 4 fundamental parameters that 1 This model is generic and could apply to all debris discs matter in this study: R0 and L? (the x and y-axes), M_ CO as long as they are made up of volatile-rich bodies and their (the point colour) , and d (the point size). On the plot, one CO content is released as gas as they are ground down within can see where the 12 systems lie, and we annotate their a steady-state collisional cascade as proposed in Matr`aet al. names, as well as the elements that have been detected so (2015). To apply the KWC16 model to a given system, we far (we omit metals as they are not expected to make up assume that the CO released is a proportion γ of the mass the bulk of the gas, Fern´andezet al. 2006). We overlay a lost through the collisional cascade. That mass loss rate can blue line showing a black body temperature of 140K. For be determined from the fractional luminosity of the debris CO adsorbed on amorphous H2O, this is the temperature disc LIR=L? and its temperature from which the planetesi- above which CO cannot be trapped in ices (under laboratory mal belt location R0 can be determined. These combine to conditions, Collings et al. 2003). Any systems to the left of give the CO input rate of M_ CO, which is one of the param- this line should not be able to retain any CO on grains (if eters in the KWC16 model, along with R0, the α viscosity, no refractories are present to hold CO).